[alsa-devel] [PATCHv2] ASoC: tlv320aic3x: Add platform data and reset gpio handling
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
---
v2:
- Removed added regulator_bulk_disable brain fart from aic3x_i2c_probe - Added comment to gpio_reset number in aic3x_pdata --- include/sound/tlv320aic3x.h | 17 +++++++++++++++++ sound/soc/codecs/tlv320aic3x.c | 25 +++++++++++++++++++++++++ 2 files changed, 42 insertions(+), 0 deletions(-) create mode 100644 include/sound/tlv320aic3x.h
diff --git a/include/sound/tlv320aic3x.h b/include/sound/tlv320aic3x.h new file mode 100644 index 0000000..b1a5f34 --- /dev/null +++ b/include/sound/tlv320aic3x.h @@ -0,0 +1,17 @@ +/* + * Platform data for Texas Instruments TLV320AIC3x codec + * + * Author: Jarkko Nikula jhnikula@gmail.com + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ +#ifndef __TLV320AIC3x_H__ +#define __TLV320AIC3x_H__ + +struct aic3x_pdata { + int gpio_reset; /* < 0 if not used */ +}; + +#endif \ No newline at end of file diff --git a/sound/soc/codecs/tlv320aic3x.c b/sound/soc/codecs/tlv320aic3x.c index b2c2794..435adf1 100644 --- a/sound/soc/codecs/tlv320aic3x.c +++ b/sound/soc/codecs/tlv320aic3x.c @@ -38,6 +38,7 @@ #include <linux/delay.h> #include <linux/pm.h> #include <linux/i2c.h> +#include <linux/gpio.h> #include <linux/regulator/consumer.h> #include <linux/platform_device.h> #include <linux/slab.h> @@ -48,6 +49,7 @@ #include <sound/soc-dapm.h> #include <sound/initval.h> #include <sound/tlv.h> +#include <sound/tlv320aic3x.h>
#include "tlv320aic3x.h"
@@ -65,6 +67,7 @@ struct aic3x_priv { struct regulator_bulk_data supplies[AIC3X_NUM_SUPPLIES]; unsigned int sysclk; int master; + int gpio_reset; };
/* @@ -1279,6 +1282,10 @@ static int aic3x_unregister(struct aic3x_priv *aic3x) snd_soc_unregister_dai(&aic3x_dai); snd_soc_unregister_codec(&aic3x->codec);
+ if (aic3x->gpio_reset >= 0) { + gpio_set_value(aic3x->gpio_reset, 0); + gpio_free(aic3x->gpio_reset); + } regulator_bulk_disable(ARRAY_SIZE(aic3x->supplies), aic3x->supplies); regulator_bulk_free(ARRAY_SIZE(aic3x->supplies), aic3x->supplies);
@@ -1303,6 +1310,7 @@ static int aic3x_i2c_probe(struct i2c_client *i2c, { struct snd_soc_codec *codec; struct aic3x_priv *aic3x; + struct aic3x_pdata *pdata = i2c->dev.platform_data; int ret, i;
aic3x = kzalloc(sizeof(struct aic3x_priv), GFP_KERNEL); @@ -1319,6 +1327,15 @@ static int aic3x_i2c_probe(struct i2c_client *i2c,
i2c_set_clientdata(i2c, aic3x);
+ aic3x->gpio_reset = -1; + if (pdata && pdata->gpio_reset >= 0) { + ret = gpio_request(pdata->gpio_reset, "tlv320aic3x reset"); + if (ret != 0) + goto err_gpio; + aic3x->gpio_reset = pdata->gpio_reset; + gpio_direction_output(aic3x->gpio_reset, 0); + } + for (i = 0; i < ARRAY_SIZE(aic3x->supplies); i++) aic3x->supplies[i].supply = aic3x_supply_names[i];
@@ -1336,11 +1353,19 @@ static int aic3x_i2c_probe(struct i2c_client *i2c, goto err_enable; }
+ if (aic3x->gpio_reset >= 0) { + udelay(1); + gpio_set_value(aic3x->gpio_reset, 1); + } + return aic3x_register(codec);
err_enable: regulator_bulk_free(ARRAY_SIZE(aic3x->supplies), aic3x->supplies); err_get: + if (aic3x->gpio_reset >= 0) + gpio_free(aic3x->gpio_reset); +err_gpio: kfree(aic3x); return ret; }
On Wed, May 05, 2010 at 01:02:03PM +0300, Jarkko Nikula wrote:
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
Acked-by: Mark Brown broonie@opensource.wolfsonmicro.com
but...
- if (aic3x->gpio_reset >= 0) {
gpio_set_value(aic3x->gpio_reset, 0);gpio_free(aic3x->gpio_reset);- }
There is gpio_is_valid().
On Wednesday 05 May 2010 13:02:03 ext Jarkko Nikula wrote:
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
v2:
- Removed added regulator_bulk_disable brain fart from aic3x_i2c_probe
- Added comment to gpio_reset number in aic3x_pdata
Would have been nice to have a commit message for this just in case. Something like this would do: Handle the reset GPIO within the codec driver in order to follow the startup protocol for the tlv320aic3x codecs.
Can you Liam add this message (if Jarkko agrees)?
I can give my: Acked-by: Peter Ujfalusi peter.ujfalusi@nokia.com Or Reviewed-by: Peter Ujfalusi peter.ujfalusi@nokia.com
Whichever you feel appropriate in case of the tlv320aic3x codec driver.
On Thu, 2010-05-06 at 11:09 +0300, Peter Ujfalusi wrote:
On Wednesday 05 May 2010 13:02:03 ext Jarkko Nikula wrote:
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
v2:
- Removed added regulator_bulk_disable brain fart from aic3x_i2c_probe
- Added comment to gpio_reset number in aic3x_pdata
Would have been nice to have a commit message for this just in case. Something like this would do: Handle the reset GPIO within the codec driver in order to follow the startup protocol for the tlv320aic3x codecs.
Can you Liam add this message (if Jarkko agrees)?
I can add a message if required.
Thanks
Liam
On Thursday 06 May 2010 11:17:55 ext Liam Girdwood wrote:
I can add a message if required.
I think it is always a good idea to explain what, and why the change. In this case the change is not that obvious.
Also all machines, which are using the tlv320aic3x should be aware, that if they do not have the reset GPIO (which they _should_), than they have to set the gpio_reset to -1, unless they want to request gpio0 in the tlv320aic3x driver for fun ;)
Thanks
Liam
On Thu, May 06, 2010 at 12:08:18PM +0300, Peter Ujfalusi wrote:
Also all machines, which are using the tlv320aic3x should be aware, that if they do not have the reset GPIO (which they _should_), than they have to set the gpio_reset to -1, unless they want to request gpio0 in the tlv320aic3x driver for fun ;)
This should be relatively safe here since the platform data was added at the same time - it's not changing the semantics underneath existing code.
On Thursday 06 May 2010 12:52:13 ext Mark Brown wrote:
On Thu, May 06, 2010 at 12:08:18PM +0300, Peter Ujfalusi wrote:
Also all machines, which are using the tlv320aic3x should be aware, that if they do not have the reset GPIO (which they _should_), than they have to set the gpio_reset to -1, unless they want to request gpio0 in the tlv320aic3x driver for fun ;)
This should be relatively safe here since the platform data was added at the same time - it's not changing the semantics underneath existing code.
Yes, it is actually _safe_
On Thu, 6 May 2010 11:09:34 +0300 Peter Ujfalusi peter.ujfalusi@nokia.com wrote:
On Wednesday 05 May 2010 13:02:03 ext Jarkko Nikula wrote:
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
v2:
- Removed added regulator_bulk_disable brain fart from aic3x_i2c_probe
- Added comment to gpio_reset number in aic3x_pdata
Would have been nice to have a commit message for this just in case. Something like this would do: Handle the reset GPIO within the codec driver in order to follow the startup protocol for the tlv320aic3x codecs.
Can you Liam add this message (if Jarkko agrees)?
Yeah, Peter is right. Even this reset 'protocol' is business as usual, reset is released after voltages are stabilized, this is kind of fix to patch adding regulator support so the commit message is nice be there.
On Wed, 2010-05-05 at 13:02 +0300, Jarkko Nikula wrote:
Signed-off-by: Jarkko Nikula jhnikula@gmail.com
v2:
- Removed added regulator_bulk_disable brain fart from aic3x_i2c_probe
- Added comment to gpio_reset number in aic3x_pdata
include/sound/tlv320aic3x.h | 17 +++++++++++++++++ sound/soc/codecs/tlv320aic3x.c | 25 +++++++++++++++++++++++++ 2 files changed, 42 insertions(+), 0 deletions(-) create mode 100644 include/sound/tlv320aic3x.h
Applied with commit message.
Thanks
Liam
participants (4)
-
Jarkko Nikula -
Liam Girdwood -
Mark Brown -
Peter Ujfalusi